EP3715496B1 - Sputtering target for producing layers containing molybdenum oxide - Google Patents

Description

The present invention relates to a sputtering target for producing molybdenum oxide-containing layers and the use of such a sputtering target for the gas phase deposition of a molybdenum oxide-containing layer.

Molybdenum oxide-containing layers (especially those based on molybdenum oxide) have interesting optical properties and are therefore particularly used in layer structures in optical or optoelectronic applications such as electronic displays. They are often used in combination with or directly adjacent to metallic layers made of, for example, aluminum (Al), copper (Cu), titanium (Ti), or (metallic) molybdenum (Mo), which function as electrical conductors, for example. An example of the application of molybdenum oxide layers is JP 2013020347 A in which metallic conductor tracks within the display of a capacitive touch sensor (touch screen) are covered by a light-absorbing layer of MoOx in order to suppress unwanted reflections of the metallic conductor tracks.

Important properties such as light absorption, light reflection, light transmission, etching rate, thermal stability and stability against other chemicals used in the manufacturing process (e.g. photoresist developer or remover) depend on the precise stoichiometric composition of the deposited molybdenum oxide-containing layer and the added doping elements (e.g. Ta, Nb, etc.). The etching rate is relevant for subsequent structuring of the deposited layer using photolithography in conjunction with a wet-chemical etching process. In many applications, such as the capacitive touch sensor mentioned above, molybdenum oxide-containing layers are required in which the oxide (molybdenum oxide and/or a molybdenum-containing mixed oxide) is present in a substoichiometric composition, i.e. the oxide has unoccupied oxygen valences and there are oxygen vacancies in the deposited molybdenum oxide-containing layer. In such substoichiometric molybdenum oxide-containing Layers (particularly based on molybdenum oxide) can achieve interesting electro-optical properties, such as high light absorption (and correspondingly low light reflection) in the visible wavelength range, particularly at a reference wavelength of 550 nm, and at the same time sufficient electrical conductivity (particularly a sheet resistance of < 20 kΩ/□ or < 20 kΩ/Square). Pure molybdenum oxide layers do have attractive electro-optical properties, but they usually have an etching rate that is too high in common wet etching media (e.g. PAN etching based on phosphoric acid, nitric acid and acetic acid for structuring aluminum metallizations, H ₂ O ₂ -based etching for structuring Cu), which leads to unstable etching results. By adding doping elements into the substoichiometric molybdenum oxide-containing layer (and correspondingly into the target material of the sputtering target), such as Ta and/or Nb, the etching rate can be reduced to an acceptable level without significantly changing the electro-optical properties.

Such molybdenum oxide-containing layers (particularly those based on molybdenum oxide) are produced industrially in a coating process using cathode sputtering ("sputtering"). In principle, it is possible to produce such molybdenum oxide-containing layers by reactive sputtering of a metallic target material, which is reactively sputtered, for example, in an appropriately adjusted argon and oxygen process gas atmosphere. The metal atoms of the target material react to a large extent only with the oxygen from the process gas during the coating process, so that the process control is made more difficult by hysteresis effects (discontinuous changes in the layer deposition rate, the discharge current and/or the target voltage due to changes in the oxygen partial pressure) and poisoning of the target surface (target poisoning). In contrast, the deposition of a sputtering target with oxidic target material is advantageous under industrial conditions, since the coating process can be carried out either with a pure noble gas process gas (usually argon) or with noble gas (usually argon) as the main component to which small amounts of other gases, such as oxygen, are added. In particular, the use of oxide target material enables higher coating rates and greater process stability while simplifying process control. Furthermore, many coating systems that are already installed are only designed for the deposition of metallic layers from a metallic sputtering target and therefore do not allow the addition of oxygen to the sputtering gas. The versatility of such sputtering systems can be increased by using an oxide target material.

The EN 10 2012 112 739 A1 describes a substoichiometric target material in which an oxygen deficiency is set either by a reduced oxide phase of substoichiometric and thus electrically conductive oxide or oxynitride based on Nb ₂ O _5-y , TiO _2-y , MoO _3-y , WO _3-y , V ₂ O _5-y (y>0) or mixtures thereof alone or by the reduced oxide phase together with a metallic admixture. In the US 2001/0020586 A1 A substoichiometric target material of chemical composition MOy is described, in which M is at least one metal from the group Ti, Nb, Ta, Mo, W, Zr and Hf.

The document EN 10 2014 111935 A1 describes a sputtering target containing a mixture of various substoichiometric metal oxides of different metals such as Mo, W, Nb, Ti, Ta, V, etc.

Apart from the above-mentioned challenges and conditions regarding molybdenum oxide-containing layers and corresponding target materials, scientific publications describe the production of powdered, molybdenum-containing mixed oxide compounds on a laboratory scale (ie product quantities of a few grams) with holding times at elevated temperatures of one or more weeks. For example, Ekström and Nygren ( Thommy Ekström and Mats Nygren, Acta Chemica Scandinavica 26 (1972) 1836-1842 ) the synthesis of (Mo _1-x Ta _x ) ₆ O ₁₄ in a homogeneity range of 0.06 ≤ x ≤ 0.08 by mixing Ta ₂ O ₅ powder with MoOs and MoO ₂ , filling the powder mixture into quartz ampoules, evacuating the quartz ampoules and heating to 640 - 750°C for a period of one to several weeks. The same publication also describes the synthesis of (Mo _1-x Nb _x ) ₅ O ₁₄ in a homogeneity range of 0.07 ≤ x ≤ 0.12. The starting powders were Nb ₂ O ₅ , MoOs and MoO ₂ , the synthesis conditions were analogous to those for (Mo _1-x Ta _x ) ₅ O ₁₄ , but the mixed oxide in a temperature range of 600 - 750 °C. Ekström and Nygren describe in another publication ( Thommy Ekström and Mats Nygren, Acta Chemica Scandinavica 26 (1972) 1827-1835 ) also the synthesis of (Mo _1-x V _x ) ₅ O ₁₄ , which was prepared from a powder mixture of V ₂ O ₅ with MoOs and MoO ₂ analogously to the mixed oxides already described, whereby a wide homogeneity range of 0.02 ≤ x ≤ 0.11 was observed when heated to 640°C or 0.05 ≤ x ≤ 0.11 when heated to 750°C. The synthesis of (Mo _1-x Ti _x ) ₅ O ₁₄ with a homogeneity range of 0.03 ≤ x ≤ 0.05 was also already described by Ekström ( Thommy Ekström, Acta Chemica Scandinavica 26 (1972) 1843-1846 ), whereby the preparation was carried out from a powder mixture of TiO ₂ , MoO ₂ and MoOs analogously to the mixed oxides already described with heating to a temperature of 600 to 750°C. In all syntheses, the product was obtained only in small quantities (gram scale) in loose powder form, not as a solid macroscopic solid.

Accordingly, the object of the present invention is to provide doped molybdenum oxide-containing sputtering targets in which a stable sputtering process can be set for producing high-quality layers, wherein the sputtering targets can be produced cost-effectively and reliably on an industrial scale.

The object is achieved by a sputtering target according to claim 1 and by a use of a sputtering target according to claim 9. Advantageous developments of the invention are specified in the dependent claims.

• von Molybdän (Mo) als metallischem Hauptbestandteil und
• von mindestens einem Dotierungselement M aus der Gruppe Tantal (Ta), Niob (Nb), Vanadium (V) und Titan (Ti).

According to the present invention, a sputtering target with an electrically conductive, oxidic target material is provided. The target material consists of at least one metal oxide

• of molybdenum (Mo) as the main metallic component and
• of at least one doping element M from the group tantalum (Ta), niobium (Nb), vanadium (V) and titanium (Ti).

The target material has - based on a micrograph of the same and measured with a Raman microscope (by means of Raman spectroscopy) - a matrix phase consisting of a mixed oxide (Mo _1-x M _x ) ₅ O ₁₄ with 0.01 ≤ x ≤ 0.13, where M in this mixed oxide is one or more elements from the group tantalum (Ta), niobium (Nb), vanadium (V) and titanium (Ti).

The present invention is based on the finding that, with regard to the process stability of the sputtering process, it is advantageous if the oxygen content of the target material is identical or essentially identical to that of the layer to be produced. The oxygen content (wt.%) of the target material and the oxygen content (wt.%) of the layer to be produced preferably differ from one another by a maximum of ± 2 wt.%. A supply of oxygen via the process gas is therefore not necessary or only required to a small extent, which suppresses hysteresis effects and simplifies the process control.

Furthermore, it must be taken into account that in principle a large number of phase compositions of a doped target material containing molybdenum oxide are possible. On the one hand, molybdenum can be present in different oxidation states, in particular in the form of MoO ₂ , of MoO ₃ , and in the form of a plurality of substoichiometric oxides MoO _y , such as Mo ₄ O ₁₁ (y=2.75), Mo ₁₇ O ₄₇ (y=2.76), Mo ₆ O ₁₄ (y=2.8), Mo ₃ O ₂₃ (y=2.875), Mo ₉ O ₂₆ (y=2.89) and Mo ₁₈ O ₅₂ (y=2.89), where Mo ₉ O ₂₆ forms the high-temperature modification and Mo ₁₈ O ₅₂ forms the low-temperature modification. According to the invention, it is provided that the at least one doping element M is contained in the target material. With regard to the possible phase compositions of the target material, this adds further degrees of freedom, because the at least one doping element M can in principle be present in elemental form in the target (ie in metallic form, e.g. as Ta, Nb, V, Ti), it can be present as an oxide (e.g. Ta ₂ O ₅ , V ₂ O ₅ , Nb ₂ O ₅ , TiO ₂ ), it can be present as a substoichiometric oxide and furthermore it can form a mixed oxide with one or more of the above-mentioned molybdenum oxides and/or with one or more oxides of the at least one further doping element M. In view of the large number of possible phase compositions, the present invention is based on the The finding is that different phases of the target material, particularly if they are partially insulating or dielectric, can lead to particle formation through the formation of arcs during sputtering. This can result in layer defects or damage to the sputtering target. Furthermore, the strength of oxide targets is rather low, especially if the microstructure has several different, unevenly distributed, coarse and/or porous phases. It is particularly critical with regard to strength if phases have different densities or CTE's (CTE: coefficient of thermal expansion) and/or adjacent phases are not sufficiently firmly connected to one another due to, for example, diffusion pores, different lattice types, etc. Different phases sometimes also have different sputtering rates, which means that the layer homogeneity and layer stoichiometry of the layers produced can be inadequate, especially if there are several unevenly distributed, coarse and porous phases of the target material. Some of the phases (e.g. MoOs) are conditionally water-soluble, which can lead to changes in the stoichiometry in the surface area and/or to the storage of residual moisture during the manufacturing process (e.g. mechanical processing, cleaning), which in turn can have a negative effect on the sputtering process and the produced layer (e.g. impairment of the vacuum, storage of H ₂ O in the layer).

In contrast, it was found that the (Mo _1-x M _x ) ₅ O ₁₄ mixed oxide phase with 0.01 ≤ x ≤ 0.13 provided according to the invention is particularly advantageous in the target material because it is electrically conductive, has no water solubility and is stable under ambient conditions. Furthermore, the doping element M is also bound (at least partially) in this matrix phase and thus in an electrically conductive phase and evenly distributed over the target material. Since this mixed oxide phase forms the matrix phase according to the invention, in which any second phases present (preferably finely distributed) can be embedded, it takes up a significant portion of the target material that passes through the sputtering target, whereby the entire target material is characterized by these advantageous properties. For example, electrical conductivity is thus ensured throughout the entire target material. and contact with water is relatively uncritical since the matrix is not attacked. As explained above, these properties are advantageous with regard to a uniform sputtering rate, the avoidance of arcs and particle formation, a homogeneous layer composition and a uniform layer thickness, stable process control during the sputtering process and with regard to simple production, storage and handling of the sputtering target. A further advantage is that the sputtering target according to the invention can be produced on an industrial scale as a compact component (compact target material of e.g. at least 1 kg per target segment) with high relative density and high phase purity within short process times (e.g. holding times typically < 12 hours), as will be described in more detail below.

"Sputtering", also known as cathode sputtering, is a physical process in which atoms or molecules are released from a solid body (target material) by bombardment with high-energy ions (e.g. noble gas ions of the process gas) and pass into the gas phase. To produce thin layers, the sputtered target material is deposited on a substrate and forms a solid layer there. This coating process, sometimes also known as sputter deposition, is one of the PVD processes (PVD: physical vapor deposition) and is also referred to below as "sputtering". The target material is the solid material of a sputtering target that is intended for sputtering during the sputtering process and for layer formation (if necessary together with other materials from the process gas and/or other sputtering targets). The sputtering target can be formed exclusively by the target material or can also have further components connected to the target material directly or indirectly (i.e. via at least one further component), such as back plate(s), support tube(s), connection piece(s), insert(s). The sputtering target (and accordingly the target material) can be provided in different geometries, in particular as a planar sputtering target with different basic shapes, such as square, rectangular, round, etc., or as a tubular sputtering target. The target material forms in particular a macroscopic solid with an extension along at least one spatial direction of at least 0.03 m (meters).

An "oxidic" target material is understood to mean that the metals contained in the target material are essentially completely present as oxides. In particular, the proportion of metallic phases in the target material based on a micrograph of the same is < 1 vol.%. In particular, the proportion of metallic phases in the target material is below the detection limit using the Raman measurements used here and described in more detail below. The target material is referred to as "electrically conductive" if it has an electrical conductivity of at least 80 S/m, measured using a resistance measuring station (Ulvac ZEM3) using the 4-point measuring method. The "phase" of the target material is understood to be a spatial region of the same within which there is a uniform chemical composition and a uniform crystal structure, whereby the orientation of the individual crystal grains within a phase can be different. A "mixed oxide" is understood to mean an oxide with a crystal structure in which the crystal lattice is composed of oxygen ions and cations of several elements (in this case in particular Mo and at least one doping element M). In this context, a "matrix phase" is understood to mean that it permeates the target material in a coherent, percolating network in which any additional phases ("second phases") are embedded as island-shaped regions (observable on the micrograph with a Raman microscope using Raman spectroscopy).

The determination of the various phases and their volume fractions as well as the density of the target material is carried out using a representative cross-section of a sample, whereby - since an isotropic structure can be assumed - the volume data are derived from the area fractions measured on the cut surface (ie the volume fractions correspond to the measured area fractions). A metallographic section of the sample is produced using dry preparation, whereby Raman spectroscopy is used with a Raman microscope for the spatially resolved determination of the various metal oxide phases. As explained below As described in detail, in Raman spectroscopy the sample surface of the target material to be analyzed is scanned point by point with a laser beam and a complete Raman spectrum is created for each measuring point ("Raman mapping"). By comparing the Raman spectrum obtained for each measuring point with reference spectra of various metal oxides in question, a corresponding phase is assigned to each measuring point and a two-dimensional representation of the phase composition of the sample is created, from which the area proportions (or the volume proportions) of the various phases can then be calculated. The (Mo _1-x M _x ) ₅ O ₁₄ matrix phase can be clearly identified using Raman spectroscopy. A reference Raman spectrum of a (Mo _0.93 Ta _0.07 ) ₅ O ₁₄ phase is shown in Figure 1 shown. The powder diffractogram data are available from the International Centre for Diffraction Data (ICDD) under PDF no. 01-070-0952. The volume data (or the identical area data) of the various metal oxide phases are standardized and relative to the total volume (or the total area) occupied by the material particles (grains) of the target material. The volume (or the identical area) occupied by the pores of the target material is excluded from this total volume (or this total area). The volume data (or the identical area data) of the individual metal oxide phases therefore add up to 100% without the pore volume.

According to the present invention, the target material can be formed from several metal oxides of different chemical composition (which then form different phases). The respective metal oxide can be a stoichiometric or a substoichiometric metal oxide of exactly one metal or alternatively of several metals (ie in the latter case a mixed oxide). In the target material, Mo forms the main component according to the invention with regard to the metals contained (which are present as cations), ie the Mo content (Mo; in at.%) relative to the sum of all metals contained (Me; in at.%) including the Mo content (Mo; in at.%), ie Mo/(Me+Mo) is at least 50%. In particular, it is at least 80%, even more preferably at least 84%. In principle, in addition to Mo and the at least one doping element M, other metals may also be contained in the target material. Preferably, however, the other metals contained in the target material in addition to Mo - apart from any impurities that may be present - are formed exclusively by the at least one doping element M. In principle, the target material can contain two or more doping elements M from the group Ta, Nb, V and Ti. In particular, however, the target material only contains exactly one doping element M from this group. Preferably, this exactly one doping element M is Ta or alternatively Nb. By adding Ta and/or Nb as a doping element, the wet etching rate in a common aluminum etchant (mixture of phosphoric acid, nitric acid, acetic acid and water) can be reduced to an acceptable level. The wet etching rate in copper etchants based on hydrogen peroxide can also be influenced by adding Ta or Nb. Ta or Ta oxide powder is a starting product that is available in large quantities, but is subject to strong price fluctuations. Nb or Nb oxide powder is usually available at a cheaper price.

As mentioned, the target material may contain manufacturing-related "impurities" (metals as well as non-metals), such as tungsten (W), sodium (Na), potassium (K), antimony (Sb), vanadium (V), chromium (Cr), iron (Fe), carbon (C) and nitrogen (N). The total content of such impurities is typically < 1,000 µg/g.

Basically, according to the present invention, the mixed oxide (Mo _1-x M _x ) ₅ O ₁₄ will contain two or more doping elements M from the group Ta, Nb, V and Ti. In particular, however, the mixed oxide (Mo _1-x M _x ) ₅ O ₁₄ will contain only one doping element M from this group. Furthermore, according to the present invention, it is basically possible for several different mixed oxides (Mo _1-x M _x ) ₅ O ₁₄ , each with M as one or more elements from the group Ta, Nb, V and Ti, to be contained in the target material, with exactly one of these mixed oxides then forming the said matrix phase. In particular, the following applies to the range x within the Mixed oxide (Mo _1-x M _x ) ₅ O ₁₄ of the matrix phase 0.02 ≤ x ≤ 0.12.

• von Molybdän (Mo) als metallischem Hauptbestandteil und
• von mindestens einem Dotierungselement M aus der Gruppe Tantal (Ta), Niob (Nb), Vanadium (V) und Titan (Ti)

gebildet. Dementsprechend gilt, dass - abgesehen von ggf. vorhandenen Verunreinigungen (Definition "Verunreinigungen" s. oberhalb) - keine weiteren Metalle enthalten sind und alle Metalle oxidisch gebunden sind (und nicht z.B. als Nitrid, Borid, etc., vorliegen). Hierdurch werden die vorteilhaften Eigenschaften von Molybdänoxid(en) in Kombination mit der Wirkung des mindestens einen Dotierungselements M (z.B. Beeinflussung der Naßätzrate) erzielt, ohne dass weitere, ggf. nachteilige Wechselwirkungen auftreten.According to the invention, the target material is formed by at least one metal oxide

• of molybdenum (Mo) as the main metallic component and
• of at least one doping element M from the group tantalum (Ta), niobium (Nb), vanadium (V) and titanium (Ti)

formed. Accordingly, apart from any impurities that may be present (see definition of "impurities" above), no other metals are contained and all metals are bound in oxide form (and not present as nitride, boride, etc.). This achieves the advantageous properties of molybdenum oxide(s) in combination with the effect of the at least one doping element M (e.g. influencing the wet etching rate) without any further, possibly disadvantageous interactions occurring.

According to the invention, the matrix phase in the target material made up of the mixed oxide (Mo _1-x M _x ) ₅ O ₁₄ has a proportion of ≥ 60 vol.%, i.e. it is in the range of 60-100 vol.%, based on a micrograph of the target material - measured with a Raman microscope (using Raman spectroscopy). In particular, the proportion is ≥ 75 vol.%, i.e. it is in the range of 75-100 vol.%, even more preferably the proportion is ≥ 90 vol.%, i.e. it is in the range of 90-100 vol.%. The higher the proportion of the mixed oxide phase, the more the target material is characterized by the advantageous properties of this matrix phase explained above. A very high proportion is only achieved if the composition of the target material approximates the ratio of the relevant elements in the mixed oxide (Mo _1-x M _x ) ₅ O ₁₄ . According to one embodiment of the invention, the proportion is 100 vol.% (ie no second phases are detectable). To achieve 100 vol.%, the composition of the target material must largely correspond to the ratio of the elements in the mixed oxide. Regarding the determination of the proportion in vol.%, reference is made to the general description above and to the detailed description below regarding sample preparation and Raman spectroscopy.

1. a. MoO₂ und Mo₄O₁₁ sowie
2. b. optional einer oder mehrerer Verbindung(en) aus der Gruppe:
   • MoO₃,
   • weiterer substöchiometrischer Molybdänoxide aus der Gruppe Mo₁₇O₄₇, Mo₅O₁₄, Mo₆O₂₃, Mo₉O₂₆ und Mo₁₈O₅₂,
   • stöchiometrischer (z.B. Ta₂O₅) und/oder substöchiometrischer (z.B. Ta₂O_5-l; 0<l<1) Ta-Oxid(e),
   • stöchiometrischer (z.B. Nb₂O₅) und/oder substöchiometrischer (z.B. Nb₂O_5-m; 0<m<1) Nb-Oxid(e),
   • stöchiometrischer (z.B. V₂O₅) und/oder substöchiometrischer (z.B. V₂O_5-n; 0<n<1) V-Oxid(e), und
   • stöchiometrischer (z.B. TiO₂) und/oder substöchiometrischer (z.B. TiO_2-o; 0<o<1) Ti-Oxid(e),

als eine oder mehrere Zweitphasen, die als Inseln in der Matrixphase verteilt sind, wobei diese Zweitphasen bezogen auf ein Schliffbild des Targetmaterials - gemessen mit einem Raman-Mikroskop (mittels Raman-Spektroskopie) - in Summe einen Anteil ≤ 40 Vol.% bilden, d.h. der Anteil liegt in dem Bereich von 0,1 - 40 Vol.%. Zweitphasen treten insbesondere dann verstärkt auf, wenn die Zusammensetzung des Targetmaterials von dem Verhältnis der Elemente in dem Mischoxid (Mo_1-xM_x)₅O₁₄ abweicht. Ist beispielsweise der Anteil an dem Dotierungselement M relativ niedrig, so dass die Bildung des Mischoxids (Mo_1-xM_x)₅O₁₄ limitiert wird, so treten Molybdänoxidphasen auf. Dabei sind unter den genannten Zweitphasen die oben genannten Phasen MoO₂ und Mo₄O₁₁ bevorzugt, da sie sie eine hohe (nahezu metallische) elektrische Leitfähigkeit (MoO₂: 1,25 × 10⁶ S/m; monoklines Mo₄O₁₁: 1,25 × 10⁶ S/m) und einen niedrigeren Dampfdruck aufweisen, was für die Stabilität des Beschichtungsprozesses vorteilhaft ist. Weitgehend entsprechende Vorteile gelten grundsätzlich auch für die optional vorhandenen Phasen aus Mo₁₇O₄₇, Mo₅O₁₄, Mo₈O₂₃, Mo₉O₂₆ und Mo₁₈O₅₂, die im Rahmen der hier betrachteten Herstellung typischerweise jedoch nicht oder nur zu geringen Anteilen auftreten. Vorzugsweise ist der Anteil der unter a. und b. genannten Zweitphasen ≤ 25 Vol.% (d.h. er liegt im Bereich 0,1 - 25 Vol.%), noch bevorzugter ≤ 10 Vol.% (d.h. er liegt im Bereich 0,1 - 10 Vol.%), so dass der verbleibende Volumenanteil jeweils durch das besonders vorteilhafte Mischoxid (Mo_1-xM_x)₅O₁₄ gebildet wird.According to the invention, the target material also consists of

1. a. MoO ₂ and Mo ₄ O ₁₁ and
2. b. optionally one or more compounds from the group:
   • _MoO3 ,
   • other substoichiometric molybdenum oxides from the group Mo ₁₇ O ₄₇ , Mo ₅ O ₁₄ , Mo ₆ O ₂₃ , Mo ₉ O ₂₆ and Mo ₁₈ O ₅₂ ,
   • stoichiometric (e.g. Ta ₂ O ₅ ) and/or substoichiometric (e.g. Ta ₂ O _5-l ; 0<l<1) Ta oxide(s),
   • stoichiometric (e.g. Nb ₂ O ₅ ) and/or substoichiometric (e.g. Nb ₂ O _5-m ; 0<m<1) Nb oxide(s),
   • stoichiometric (e.g. V ₂ O ₅ ) and/or substoichiometric (e.g. V ₂ O _5-n ; 0<n<1) V oxide(s), and
   • stoichiometric (e.g. TiO ₂ ) and/or substoichiometric (e.g. TiO _2-o ; 0<o<1) Ti oxide(s),

as one or more second phases that are distributed as islands in the matrix phase, whereby these second phases, based on a micrograph of the target material - measured with a Raman microscope (using Raman spectroscopy) - make up a total proportion of ≤ 40 vol.%, i.e. the proportion is in the range of 0.1 - 40 vol.%. Second phases occur particularly frequently when the composition of the target material deviates from the ratio of the elements in the mixed oxide (Mo _1-x M _x ) ₅ O _14. If, for example, the proportion of the doping element M is relatively low, so that the formation of the mixed oxide (Mo _1-x M _x ) ₅ O ₁₄ is limited, molybdenum oxide phases occur. Of the second phases mentioned, the above-mentioned phases MoO ₂ and Mo ₄ O ₁₁ are preferred because they have a high (almost metallic) electrical conductivity (MoO ₂ : 1.25 × 10 ⁶ S/m; monoclinic Mo ₄ O ₁₁ : 1.25 × 10 ⁶ S/m) and a lower vapor pressure, which is advantageous for the stability of the coating process. Largely corresponding advantages also apply in principle to the optionally present phases of Mo ₁₇ O ₄₇ , Mo ₅ O ₁₄ , Mo ₈ O ₂₃ , Mo ₉ O ₂₆ and Mo ₁₈ O ₅₂ , which, however, typically do not occur or only occur in small proportions in the context of the production considered here. The proportion of the phases under a. and b. is preferably said second phases ≤ 25 vol.% (ie it is in the range 0.1 - 25 vol.%), more preferably ≤ 10 vol.% (ie it is in the range 0.1 - 10 vol.%), so that the remaining volume fraction is formed by the particularly advantageous mixed oxide (Mo _1-x M _x ) ₅ O ₁₄ .

In the target material, based on a micrograph of the same - measured with a Raman microscope (using Raman spectroscopy) - the proportion of the second phase MoOs is in the range of 0 - 2 vol.%, i.e. the proportion is ≤ 2 vol.%. The proportion of the second phase MoOs should be kept as low as possible, since MoOs is electrically insulating (electrical conductivity less than 1 × 10 ^-5 S/m), which can lead to particle formation during the coating process. In addition, MoOs is water-soluble, which is disadvantageous during mechanical processing and storage of the target. Preferably, a second phase MoOs in the target material is not detectable at all using Raman spectroscopy (i.e. measured proportion of 0 vol.%), or - if it cannot be completely avoided - is kept to a proportion in the range of 0.1 - 1.0 vol.%.

In the target material according to the invention, the oxygen content in wt.% is in the range from 26 to 32 wt.%. The oxygen content in the deposited layer influences the optical reflection, so that a largely corresponding oxygen content, which is typically in the above-mentioned range, is preferably set in the target material. Furthermore, sufficient electrical conductivity is achieved within this range, so that a DC (direct current) sputtering process is possible (while electrically insulating targets have to be sputtered with an RF (RF: radio frequency; i.e. high frequency) sputtering process (i.e. alternating current), which is associated with increased costs (particularly with regard to the design of the sputtering system). The oxygen content can - as described in more detail below - be adjusted during production by weighing in appropriate oxide-containing powders, so that the weighed-in oxygen content (in relation to the weighed-in metal content) corresponds to the desired oxygen content of the target material. At the same time, the oxygen content mentioned is favorable for the sintering behavior and the formation of the mixed oxide (Mo _1-x M _x ) ₅ O ₁₄ during production. It is useful to consider only the total oxygen content in the target material, since the exact occupancy of the various lattice sites in the metal oxide phases can also change depending on the temperature. The total oxygen content in the target material is determined by carrier gas hot extraction (Detection of oxygen as CO or CO ₂ using an infrared measuring cell, e.g. with a LECO RO300 or LECO 836 device).

According to the invention, the proportion of the (at least one) doping element M in at.% in the target material relative to the total content of Mo and the doping element M in at.% in total, ie M/(Mo+M), is in the range of 0.02 - 0.15. A minimum content of the (at least one) doping element M is required so that the mixed oxide (Mo _1-x M _x ) ₅ O ₁₄ can form as a matrix phase in the target material. There is a certain amount of leeway here, since the proportion x of the doping element M within the matrix phase (Mo _1-x M _x ) ₅ O ₁₄ can vary to a certain extent and can therefore also be slightly reduced. Furthermore, a minimum content of the doping element M is required so that the desired properties of the layer produced (in particular with regard to wet etching behavior) are present. Conversely, too high a content of the doping element M is disadvantageous because the doping element M can then no longer be absorbed in the matrix phase and consequently leads to an increased proportion of the oxide of the doping element (e.g. Ta ₂ O ₅ , Nb ₂ O ₅ , V ₂ O ₅ , TiO ₂ ) in the form of a second phase. Furthermore, the reflection of the layers produced increases again with increasing content of the doping element M (e.g. Ta) and the etching rate no longer corresponds to the desired range. Against this background, the range of 0.05 - 0.12 for the ratio M/(Mo+M) is particularly preferred. The desired ratio M/(Mo+M) can be set during production by weighing out corresponding powders, since this remains largely unchanged with the production route described below. The proportions of the metals contained in the target material itself as well as the ratio M/(Mo+M) can be determined by chemical analysis, in particular by ICP-MS (Inductively Coupled Plasma Mass Spectroscopy) or ICP-OES (Inductively Coupled Plasma Optical Emission Spectrometry).

According to a further development, the doping element M is formed by Ta in the target material and the matrix phase of the mixed oxide is formed by (Mo _1-x Ta _x ) ₆ O ₁₄ with 0.06 ≤ x ≤ 0.08. Typically, the Ta content x in the matrix phase (Mo _1-x Ta _x ) ₆ O ₁₄ has the value 0.07. Depending on the availability of Ta to form the matrix phase during production (i.e. in particular depending on the composition and quantity ratio of the starting powders), x is generally within the range 0.06 ≤ x ≤ 0.08, with the matrix phase being formed stably in this range. As mentioned above, Ta is particularly preferred with regard to influencing the wet etching rate and availability. The matrix phase (Mo _1-x Ta _x ) ₆ O ₁₄ has also proven to be particularly advantageous in practice.

Preferably, as high a proportion as possible of the added doping element M (here: Ta) is incorporated into the matrix phase of the mixed oxide (Mo _1-x M _x ) ₅ O ₁₄ and preferably the doping element M (here: Ta) - if it is not completely incorporated - is present as a metal oxide (e.g. mixed oxide, Ta ₂ O ₅ ) and not as a metallic element. Ta as a metallic phase would lead to significantly different properties. The proportion of Ta ₂ O ₅ is also preferably relatively low, since Ta ₂ O ₅ is dielectric and can therefore be the cause of particles and layer defects during sputtering. Furthermore, Ta ₂ O ₅ would be optically transparent as a thin partial layer in the produced layer and would therefore impair the optical properties of the layer. Accordingly, according to a further development, Ta (preferably as the only doping element M contained) is present in the target material entirely as a metal oxide (e.g. mixed oxide, Ta ₂ O ₅ ) (i.e. not as a metallic element), wherein in the target material, based on a micrograph of the same - measured with a Raman microscope (by means of Raman spectroscopy) - the proportion of the phase Ta ₂ O ₅ is ≤ 3 vol.% (i.e. proportion of 0-3 vol.%), preferably ≤ 1 vol.% (i.e. proportion of 0-1 vol.%). This applies in particular to a proportion of the doping element Ta in at.% relative to the total content of Mo and Ta in at.%, ie M/(Mo+M), in the range from 0.02 - 0.15, preferably in the range from 0.05 - 0.12.

According to a further development, the doping element M in the target material is formed by Nb and the matrix phase of the mixed oxide is formed by (Mo _1-x Nb _x ) ₅ O ₁₄ with 0.07 ≤ x ≤ 0.12. Depending on the availability of Nb for the formation of the matrix phase during production (ie in particular depending on Composition and quantity ratio of the starting powders) x is usually within the range of 0.07 ≤ x ≤ 0.12, whereby the matrix phase is formed stably in this range. As mentioned above, Nb is advantageous in terms of influencing the wet etching rate and low-cost availability, also because the matrix phase is formed stably over a wider range of 0.07 ≤ x ≤ 0.12.

As explained above with regard to Ta, preferably as high a proportion as possible of the added doping element M (here: Nb) is incorporated into the matrix phase of the mixed oxide (Mo _1-x M _x ) ₅ O ₁₄ and preferably the doping element M (here: Nb) - if it is not completely incorporated - is present as a metal oxide (e.g. mixed oxide, Nb ₂ O ₅ ) and not as a metallic element. Accordingly, according to a further development, Nb (preferably as the only doping element M contained) is present in the target material entirely as a metal oxide (e.g. mixed oxide, Nb ₂ O ₅ ) (ie not as a metallic element), wherein in the target material, based on a micrograph of the same - measured with a Raman microscope (by means of Raman spectroscopy) - the proportion of the phase Nb ₂ O ₅ is ≤ 3 vol.% (ie proportion of 0-3 vol.%), preferably ≤ 1 vol.% (ie proportion of 0-1 vol.%). This applies in particular to a proportion of the doping element Nb in at.% relative to the total content of Mo and Nb in at.%, ie M/(Mo+M), in the range of 0.02 - 0.15, preferably in the range of 0.05-0.12.

According to a further development, the target material has a relative density of ≥ 95% (i.e. in a range of 95 - 99.9%), in particular of ≥ 97% (i.e. in a range of 97 - 99.9%), preferably of ≥ 98% (i.e. in a range of 98 - 99.9%). A compact target material with a high relative density is important for the quality of the deposited layers, since less dense target materials can lead to lightning discharges ("Ares") and particle formation during sputtering. The relative density is determined by means of digital image analysis based on light microscopic images of the metallographic section of the target material, in which the relative surface area "FA" of the pores (FA: surface area of the pores relative to the total area examined) is evaluated. The relative density then corresponds to the value (1-FA), where it is calculated as the arithmetic mean of three such porosity measurements. The procedure is described in detail below.

According to a further development, the target material has an electrical conductivity of ≥ 100 S/m at 25°C. This enables DC sputtering and sufficient electrical conductivity is also achieved in the produced layer. The electrical conductivity is measured using the 4-point measuring method via a resistance measuring station (e.g. Ulvac ZEM3).

According to a further development, the target material is in the form of a (macroscopic) solid produced by powder metallurgy. The powder metallurgy production route, which is not according to the invention, enables cost-effective production of the target material on an industrial scale as a compact component (compact target material of e.g. at least 1 kg per target segment) with high relative density and high phase purity within short process times (e.g. holding times typically < 12 hours). Powder metallurgical production is understood to mean that corresponding starting powders (which, in the case of several powders, are mixed beforehand) are compacted by applying pressure and/or temperature. The powder metallurgical production leads to a characteristic microstructure of the target material, in particular to a multi-phase and fine-grained structure. The powder metallurgical production is easily recognizable by a person skilled in the art on the basis of a micrograph of the target material in a light microscope, in a scanning electron microscope (SEM) or via a Raman microscope.

The starting powder is preferably a powder mixture of MoO ₂ , MoO ₃ , optionally containing small amounts of substoichiometric molybdenum oxides such as in particular Mo ₄ O ₁₁ , and one or more oxides of the doping element M (Ta ₂ O ₅ , V ₂ O ₅ , Nb ₂ O ₅ , TiO ₂ ). The amounts of the various powders are weighed in such a way that the desired ratio of the elements Mo, M and O is set. MoO ₂ and MoOs are easily accessible, inexpensive and thermodynamically stable raw materials under ambient conditions, easy to handle. Substoichiometric oxides can by reduction of MoOs powder in an appropriate atmosphere such as H ₂ .

In order to achieve the formation of the desired matrix phase from the mixed oxide (Mo _1-x M _x ) ₅ O ₁₄ during the subsequent compaction step with the desired high proportion, it is important to ensure that the proportion of the weighed powders is close to that of the mixed oxide. It is also important that the powders used are very fine-grained (with d50 values typically < 10 µm) and are mixed very well before the compaction step. As is also explained using the production examples, the coarse fraction > 63 µm is sieved off in the MoO ₂ powder and the powder from the oxide of the doping element M (and possibly also in the substoichiometric molybdenum oxides) using an appropriate sieving step (e.g. sieve with a mesh size of 63 µm). A particularly fine powder is also used for the MoOs powder (e.g. with a d50 value of 3 µm, particle size measurement by laser diffraction, e.g. using a Malvern device). The powders are preferably mixed in an intensive mixer or with a plowshare mixer, whereby care must be taken to ensure that the powders are very well mixed.

Compaction can be carried out in particular by means of hot pressing, hot isostatic pressing, spark plasma sintering (SPS) or press sintering. Compaction takes place in particular at temperatures between 600 and 800 °C and pressing pressures between 15 and 110 MPa. Compaction preferably takes place in a vacuum or in a protective gas atmosphere (e.g. argon), especially if elevated temperatures are set. With SPS, compaction takes place by applying pressure and temperature, with the heat being generated internally by an electric current passed through the powder mixture. With hot pressing, compaction also takes place by applying pressure and temperature, with the heat being supplied from outside via a heated mold. With hot isostatic pressing, the starting powder is in a closed capsule and compaction takes place by applying pressure and temperature to the capsule. With compaction by means of press sintering, the starting powder is pressed into a green compact. and then sintered by heat treatment below the melting temperature.

During the compaction process, the starting powders are converted into the target material in a solid phase reaction or, depending on the chemical composition and process conditions, also in a liquid phase reaction or multi-phase reaction (e.g. solid-liquid). The reactions that take place are similar to a comproportionation: MoOs is reduced to possibly different substoichiometric oxides (e.g. Mo ₁₈ O ₅₂ , MoO ₄ O ₁₁ , ...), while MoO ₂ is at least partially oxidized to possibly different substoichiometric oxides. Furthermore, at least one oxide of the doping element M (Ta ₂ O ₅ , V ₂ O ₅ , Nb ₂ O ₅ , TiO ₂ ) is incorporated to a high extent into the mixed oxide (Mo _1-x M _x ) ₅ O ₁₄ to form the matrix phase. After compaction, mechanical post-processing can be carried out, for example using cutting tools, to achieve the desired final geometry or surface preparation (setting the desired surface roughness). If necessary, further components such as back plate(s), support tube(s), connection piece(s), insert(s) must be attached to complete the sputtering target.

According to a further development of the present invention, in the target material, based on a micrograph of the same - measured with a Raman microscope (using Raman spectroscopy) - one or more second phases are formed in the matrix phase as finely and evenly distributed, small islands. In particular, at least 95% of the islands of the second phase(s) have an equivalent circular diameter (each calculated from the relevant area of the island, shown in the form of a circle with the same area and with an "equivalent circular diameter") of ≤ 60 µm each. The number of "small islands with an equivalent circular diameter of ≤ 60 µm each" is considered relative to the total number of islands in a representatively selected, analyzed area of 1,000 µm × 1,000 µm. This fine island structure of the second phase(s) can be visualized with a Raman image of the microsection and can be evaluated accordingly by means of quantitative image analysis with regard to the area and size of the islands.

The present invention further relates to the use of a sputtering target according to the present invention, which can additionally be designed according to one or more of the developments and variants described above, for the gas phase deposition of a molybdenum oxide-containing layer, wherein the sputtering process is carried out as a DC sputtering process or pulsed DC sputtering process in a noble gas atmosphere without oxygen or alternatively with the supply of at most 20 vol.% oxygen as a reactive gas. In direct current sputtering, or DC sputtering, a direct voltage is applied between the sputtering target connected as a cathode and an anode (usually the housing of the coating system and/or shielding plates in the vacuum chamber). The DC sputtering process or pulsed DC sputtering process takes place in a noble gas atmosphere, in particular an argon gas atmosphere, preferably non-reactively without the additional supply of oxygen. The layers deposited in this way may have a slightly lower oxygen content than the target material used due to a slight oxygen depletion during the coating process (removal of oxygen formed via the extracted process gas), although this can vary slightly depending on the coating system and operating mode. In order to produce molybdenum oxide-containing layers with a higher oxygen content than that of the target material, the target material can also be reactively sputtered with the addition of a maximum of 20 vol.% oxygen (based on the composition of the reactive gas), preferably < 10 vol.%, even more preferably < 5 vol.%, whereby the amount of oxygen supplied can be kept comparatively low by using an oxidic target material. The disadvantages of reactive sputtering (hysteresis effects, potential inhomogeneities of the deposited layer) are therefore not as pronounced.

Further advantages and benefits of the invention will become apparent from the following description of embodiments with reference to the attached figures. In the figures showing Raman spectra, the intensity in (number or counts) is plotted against the Raman shift in (cm ^-1 ) (Raman shift).

Fig. 1:

Raman-Referenzspektrum des Mischoxids (Mo_0.93Ta_0.007)₅O₁₄;

Fig. 2:

Raman-Referenzspektrum von Ta₂O₅;

Fig. 3:

Raman-Referenzspektrum von Mo₁₈O₅₂;

Fig. 4:

Raman-Referenzspektrum von MoO₂;

Fig. 5:

Raman-Referenzspektrum von MoOs;

Fig. 6:

Raman-Referenzspektrum von Mo₄O₁₁;

Fig. 7:

eine mittels Raman-Mapping erstellte Mikrostruktur eines erfindungsgemäßen Ausführungsbeispiels;

Fig. 8:

eine mittels Raman-Mapping erstellte Mikrostruktur eines Vergleichsbeispiels;

Fig. 9:

ein Diagramm, in dem die Abscheideraten jeweils in (nm/min), d.h. in Nanometer/Minute, für zwei erfindungsgemäße Sputteringtargets (Bsp. 1, Bsp. 2) und ein Vergleichsbeispiel (Referenz) aufgetragen sind (jeweils kurz als MoO_z (2 < z < 3) mit dem entsprechenden molaren Anteil an Ta₂O₅, jeweils bezogen auf die Ausgangspulver, bezeichnet); und

Fig. 10:

ein Diagramm, in dem die Reflektivität in (%) über der Wellenlänge (nm) (Nanometer) aufgetragen ist, für Schichtsysteme, die mit den in Fig. 8 dargestellten zwei erfindungsgemäßen Sputteringtargets (Bsp. 1, Bsp. 2) und dem Vergleichsbeispiel (Referenz) abgeschieden wurden.

From the figures show:

Fig.1:

Raman reference spectrum of the mixed oxide (Mo _0.93 Ta _0.007 ) ₅ O ₁₄ ;

Fig. 2:

Raman reference spectrum of Ta ₂ O ₅ ;

Fig. 3:

Raman reference spectrum of Mo ₁₈ O ₅₂ ;

Fig.4:

Raman reference spectrum of MoO ₂ ;

Fig.5:

Raman reference spectrum of MoOs;

Fig.6:

Raman reference spectrum of Mo ₄ O ₁₁ ;

Fig.7:

a microstructure of an embodiment according to the invention created by means of Raman mapping;

Fig.8:

a microstructure of a comparative example created by Raman mapping;

Fig. 9:

a diagram in which the deposition rates are plotted in (nm/min), ie in nanometers/minute, for two sputtering targets according to the invention (Example 1, Example 2) and a comparative example (reference) (each referred to briefly as MoO _z (2 < z < 3) with the corresponding molar proportion of Ta ₂ O ₅ , each based on the starting powder); and

Fig. 10:

a diagram in which the reflectivity in (%) is plotted against the wavelength (nm) (nanometers) for layer systems that are made with the Fig.8 shown two sputtering targets according to the invention (Ex. 1, Ex. 2) and the comparative example (reference).

Example 1:

MoO ₂ powder (Plansee SE; d50 value of approx. 6 µm) is sieved in a sieve with a mesh size of 63 µm, whereby the coarse fraction >63 µm is not reused. 40 mol.% of the MoO ₂ powder obtained in this way is mixed with 57 mol.% MoOs powder (Molymet, present with a d50 value of approx. 3 µm) and 3 mol.% tantalum pentoxide powder (HC Starck, d50 value of approx. 2 µm; sieved to <63 µm; Ta ₂ O ₅ ) in a plowshare mixer (Lödige) for 20 minutes, so that a homogeneous distribution between the powder components is achieved.

The resulting powder mixture is placed in a graphite mold with dimensions of 0250 mm (i.e. diameter of 250 mm) and a height of 50 mm and compacted in a hot press under vacuum at a pressure of 30 MPa, a temperature of 800 °C and a holding time of 480 minutes. The compacted component has a relative density of 97.5%. The majority of the resulting target material consists of the matrix phase of the mixed oxide (Mo _0.93 Ta _0.07 ) ₅ O ₁₄ with a proportion of 79.4 vol.%. The target material also has a MoO ₂ phase with a proportion of 8.5 vol.% and a Mo ₄ O ₁₁ phase with a proportion of 12.1 vol.%.

Example 2:

MoO ₂ powder (Plansee SE; d50 value of approx. 6 µm) is sieved in a sieve with a mesh size of 63 µm, whereby the coarse fraction >63 µm is not reused. 40 mol.% of the MoO ₂ powder obtained in this way is mixed with 57 mol.% MoOs powder (Molymet, in this case with a d50 value of approx. 3 µm) and 3 mol.% tantalum pentoxide powder (HC Starck, d50 value of approx. 2 µm; sieved to <63 µm; Ta ₂ O ₅ ) in an intensive mixer (Eirich) for 10 minutes, so that a homogeneous distribution between the powder components is achieved. The resulting powder mixture is placed in a graphite mold with dimensions of 0250 mm (i.e. diameter of 250 mm) and a height of 50 mm and compacted in a hot press under vacuum at a pressure of 30 MPa, a temperature of 800 °C and a holding time of 360 minutes. The compacted component has a relative density of 99.5%. The majority of the target material obtained consists of the matrix phase of the mixed oxide (Mo _0.93 Ta _0.07 ) ₅ O ₁₄ with a proportion of 94.3 vol.%. The target material also has a MoO ₂ phase with a proportion of 2.6 vol.%, a Mo ₄ O ₁₁ phase with a proportion of 3.0 vol.% and a Ta ₂ O ₅ phase with a proportion of 0.1 vol.%.

In Fig.7 The microstructure of the target material produced in this way, determined by Raman mapping, is shown. In the microstructure, areas with Mo0 ₂ phase, areas with Mo ₄ O ₁₁ phase and areas with Ta ₂ O ₅ phase can be seen, corresponding to the phase proportions shown above. These different Phases are embedded in a matrix which is formed by the (Moo. ₉₃ Tao.o ₇ ) ₅ 0i ₄ phase.

In contrast, Fig.8 As a comparative example, a Raman mapping of a target material is shown which does not have a matrix phase of the mixed oxide (Mo _0.93 Ta _0.07 ) ₅ O ₁₄ . The (Mo _0.93 Ta _0.07 ) ₅ O ₁₄ phase is present in a proportion of 24.0 vol.%. The Mo ₄ O ₁₁ phase accounts for 50.3 vol.%, the MoO ₂ phase for 12.5 vol.%, the Mo ₁₈ O ₅₂ phase for 5.0 vol.%, the Ta ₂ O ₃ phase for 2.4 vol.% and the MoOs phase for 5.8 vol.%. As can also be seen from Fig.8 As can be seen, the phases within the target material of the comparative example are significantly coarser and more unevenly distributed compared to the target material according to the invention from Fig.7 .

Example 3:

MoO ₂ powder (Plansee SE; d50 value of approx. 6 µm) is sieved in a sieve with a mesh size of 63 µm, whereby the coarse fraction >63 µm is not reused. 52 mol.% of the MoO ₂ powder obtained in this way is mixed with 42 mol.% MoOs powder (Molymet, present with a d50 value of approx. 3 µm) and 6 mol.% tantalum pentoxide powder (HC Starck, d50 value of approx. 2 µm; sieved to <63 µm; Ta ₂ O ₅ ) in a plowshare mixer (Lödige) for 20 minutes, so that a homogeneous distribution between the powder components is achieved. The resulting powder mixture is placed in a graphite mold with dimensions of 0250 mm (i.e. diameter of 250 mm) and a height of 50 mm and compacted in a hot press under vacuum at a pressure of 30 MPa, a temperature of 800 °C and a holding time of 480 minutes. The compacted component has a relative density of 97.5%. The majority of the resulting target material consists of the matrix phase of the mixed oxide (Mo _0.93 Ta _0.07 ) ₅ O ₁₄ with a proportion of 80.3 vol.%. The resulting target material has a MoO ₂ phase with a proportion of 18.9 vol.%, a Mo ₄ O ₁₁ phase with a proportion of 0.2 vol.% and a Ta ₂ O ₅ phase with a proportion of 0.6 vol.%.

Example 4:

MoO ₂ powder (Plansee SE; d50 value of approx. 6 µm) is sieved in a sieve with a mesh size of 63 µm, whereby the coarse fraction >63 µm is not reused. 45 mol.% of the MoO ₂ powder obtained in this way is mixed with 51 mol.% MoOs powder (Molymet, in this case with a d50 value of approx. 3 µm) and 4 mol.% niobium pentoxide powder (HC Starck, d50 value of approx. 2 µm; sieved to <63 µm; Nb ₂ O ₅ ) in a plowshare mixer (Lödige) for 20 minutes, so that a homogeneous distribution between the powder components is achieved. The resulting powder mixture is placed in a graphite mold with dimensions of 0250 mm (i.e. diameter of 250 mm) and a height of 50 mm and compacted in a hot press under vacuum at a pressure of 30 MPa, a temperature of 800 °C and a holding time of 360 minutes. The majority of the resulting target material consists of the matrix phase of the mixed oxide (Mo _0.91 Nb _0.09 ) ₅ O ₁₄ phase with a proportion of 78.6 vol.%. The compacted component has a relative density of 98.2%. The resulting target material has a MoO ₂ phase with a proportion of 18.7 vol.%, a Mo ₄ O ₁₁ phase with a proportion of 2.6 vol.% and a Nb ₂ O ₃ phase with a proportion of 0.1 vol.%.

Example 5:

MoO ₂ powder (Plansee SE; d50 value of approx. 6 µm) is sieved in a sieve with a mesh size of 63 µm, whereby the coarse fraction >63 µm is not reused. 39 mol.% of the MoO ₂ powder obtained in this way is ground with 52 mol.% MoOs powder (Molymet, in this case with a d50 value of approx. 3 µm) and 8 mol.% titanium dioxide powder (rutile; d50 value of < 1 µm; TiO ₂ ) in a ball mill for 10 minutes and then mixed in an intensive mixer (Eirich) for 10 minutes to achieve a homogeneous distribution between the powder components. The resulting powder mixture is placed in a graphite mold with dimensions of Ø100mm and a height of 50 mm and compacted in a hot press under vacuum at a pressure of 30 MPa, a temperature of 800 °C and a holding time of 120 min. The compacted The component has a relative density of 91.2%. The target material obtained has a matrix phase of the mixed oxide (Mo _0.97 Ti _0.03 ) ₅ O ₁₄ .

Sputtering tests:

As part of a series of tests, the molybdenum-tantalum oxide target materials produced according to Example 1 and Example 2 were sputtered non-reactively under specified process conditions in comparison with a non-inventive molybdenum-tantalum oxide target material (reference) in order to determine the layer properties. The molybdenum-tantalum oxide target material used as a reference had the same composition as in Examples 1 and 2, but without a matrix phase made of the mixed oxide (Mo _0.93 Ta _0.07 ) ₅ O ₁₄ being present. A sputtering power of 100 W and an argon process pressure of 5.0 × 10 ^-3 mbar (22 sccm; sscm: standard cubic centimeters per minute) were used. The deposition rate of the sputtering targets used is in Fig.9 shown, with the left column in Fig.9 the target material from Example 1 (Ex. 1), the middle column corresponds to the target material from Example 2 (Ex. 2) and the right column corresponds to the target material of the reference.

The reflectivity of the layers produced was used as a criterion for the assessment. To determine the reflectivity, glass substrates (Corning Eagle XG, 50 × 50 × 0.7 mm ³ ) were coated with molybdenum-tantalum oxide and a cover layer of 200 nm Al (aluminum). The molybdenum-tantalum oxide target materials mentioned above, produced according to Example 1 and Example 2, were sputtered non-reactively under specified process conditions in comparison with the above-mentioned non-inventive molybdenum-tantalum oxide target material (reference). The reflection was measured through the glass substrate using a Perkin Elmer Lambda 950 photospectrometer over the specified wavelength range. In order to achieve the lowest possible reflectivity, the layer thickness of the molybdenum-tantalum oxide was varied in a range from 40 to 60 nm. The results from this series of tests for a molybdenum-tantalum oxide layer thickness of 50 nm are shown in Fig.10 where the reflectivity in (%) is plotted against the wavelength in (nm; nanometers). Reflectivity curves from example 1 (example 1; dotted curve) and example 2 (example 2; dashed curve) almost lie on top of each other, except that the reflectivity curve of example 1 is slightly increased in the range of higher wavelengths. The curves lie significantly below the solid reflectivity curve of the non-inventive reference.

As can be seen from a comparison of the Figs. 9 and 10 As can be seen, the target materials according to Examples 1 and 2 achieve a lower reflectivity over the wavelength range considered (see Fig.10 ), and this with comparable separation rates (see Fig.9 ). Furthermore, the target materials according to the invention have the other advantages explained above. In particular, due to their electrical conductivity and advantageous phase composition (matrix phase; fine distribution of the second phases), they have a preferred sputtering behavior (uniform sputtering rate while avoiding arcs and particle formation), which has a positive effect on the layer quality (homogeneous layer composition; uniform layer thickness).

Sample preparation:

To determine the volume fractions of the phases contained in the target material and the density of the target material, a metallographic section was produced from a representative sample part using dry preparation. A sample with an area of approx. 10-15 × 10-15 mm ² was cut using a dry cutting method (diamond wire saw, band saw, etc.), cleaned using compressed air, then embedded in phenolic resin while warm and conductive (C-doped), ground and polished. Since at least the MoOs phase portion is water-soluble, dry preparation is important. The section obtained in this way was then analyzed using a light microscope.

Determination of phase proportions using Raman spectroscopy:

For the spatially resolved determination of the phases contained in the target material, a Raman microscope (Horiba LabRAM HR800) was used, in which a confocal light microscope (Olympus BX41) was combined with a Raman spectrometer The surface to be analyzed was scanned point by point with a step size of 5 µm on an area of 1 × 1 mm ² using a focused laser beam (He-Ne laser, wavelength λ = 632.81 nm, 15 mW total power) (the sample surface to be analyzed was fixed to a motorized XYZ table and this was moved). A complete Raman spectrum was created for each of the 201 × 201 measuring points ("Raman mapping"). Raman spectra are obtained from the backscattered radiation, which is split wavelength-dispersively by an optical grating (300 lines/mm; spectral resolution: 2.6 cm ^-1 ) and recorded using a CCD detector (1024 × 256 pixel multichannel CCD; spectral range: 200-1050 nm). With a microscope objective with 100x magnification and numerical aperture NA of 0.9, which is used to focus the laser beam from the Raman spectrometer, a theoretical measuring point size of 0.7 µm ² was achieved. The excitation energy density (5 mW/µm ² ) is chosen to be low enough to avoid phase transformations in the sample. The penetration depth of the excitation radiation is limited to a few micrometers for molybdenum oxides (in the case of pure MoOs, for example, to approx. 4 µm). For each measuring point, the Raman signal was averaged over 1 s (s: second) acquisition time, resulting in a sufficiently good signal-to-noise ratio. By automated evaluation of these Raman spectra (Horiba LabSpec 6 evaluation software), a two-dimensional representation of the surface composition of the sample was created, from which the domain size, area proportions, etc. of the various phases can be determined quantitatively. To precisely identify a molybdenum oxide phase, reference spectra are recorded on previously synthesized reference samples or reference spectra on larger homogeneous sample areas, whereby care is taken to ensure that a reference spectrum corresponds exactly to one metal oxide phase. In the Fig. 1 to 6 are used reference spectra of (Mo _0.93 Ta _0.07 ) ₅ O ₁₄ ( Fig.1 ), Ta ₂ O ₅ ( Fig.2 ), Mon ₁₈ O ₅₂ ( Fig.3 ), _MoO2 ( Fig.4 ), MoOs ( Fig.5 ), Mon ₄ O ₁₁ ( Fig.6 ) (in the individual spectra, the intensity (number) of scattered light is plotted against the Raman shift (cm ^-1 )). The analysis and assignment of the Raman spectra is carried out using the "Multivariate Analysis Module" of the above-mentioned evaluation software using the CLS method (classical least square fitting). The sample spectrum S is used as a linear combination of the individual normalized reference spectra R _i , where c _i is the respective weighting factor and Δ is an offset value, S= Σc _i R _i + Δ. Each measurement point is then assigned a color that corresponds to a metal oxide phase, whereby only the phase with the highest weighting factor c _i is used for the color assignment. The size (the amount) of the weighting factor c _i determines the brightness of the measurement point. This procedure is justified insofar as the spectrum of a measurement point can usually be clearly assigned to a single metal oxide phase.

With the lens used, a sample spectrum was obtained from all 201 × 201 measuring points, even when measurements were taken on a pore. In this case, the signal came from a deeper area below the pore. If no Raman spectrum is obtained for individual measuring points, e.g. due to a pore, this can be excluded from the determination of the area proportions, i.e. the volume taken up by the pores of the target material is excluded from the total volume. The volume data for the individual molybdenum oxide phases therefore add up to 100% without the pore volume.

The analytical method described here is very well suited for determining the relative phase proportions of the phases in question.

Determination of relative density:

The relative density of the target material is determined by digital image analysis of light microscopic images of the metallographic section, in which the relative surface area "FA" of the pores is determined. After preparation of the samples, three bright field images were taken each with a size of 1 × 1mm ² at 100x magnification, whereby zones of obvious breakouts or other damage patterns such as scratches due to dry preparation were avoided where possible. The images obtained were evaluated using the digital image processing software implemented in the IMAGIC image database. For this purpose, the pore proportion (dark) was determined on the image based on a histogram depending on the gray level. marked. The lower limit of the interval was set at 0 (= black). In contrast, the upper limit must be subjectively estimated using the grayscale intensity histogram (255 = white). The image area to be measured was set (ROI) to exclude the scale bar. The result is the relative area share "FA" of the pores (in percent) and the image colored according to the selected grayscale interval (colored means that this pixel was included in the measurement and was therefore counted as a pore). The value (1-FA) for the relative density was determined as the arithmetic mean of three such porosity measurements.

Claims (9)

1. Sputtering target comprising an electrically conductive, oxidic target material, where the target material consists, together with impurities, of at least one metal oxide

   - of molybdenum (Mo) as metallic main constituent, and of at least one doping element M from the group consisting of tantalum (Ta), niobium (Nb), vanadium (V) and titanium (Ti), characterized in that the target material comprises, based on a polished section thereof - measured using a Raman microscope as indicated in the description - a matrix phase composed of a mixed oxide (Mo_1-xM_x)₅O₁₄ where 0.01 ≤ x ≤ 0.13, where M in this mixed oxide is one or more elements from the group consisting of tantalum (Ta), niobium (Nb), vanadium (V) and titanium (Ti), wherein the matrix phase composed of the mixed oxide in the target material has, based on a polished section of the target material - measured using a Raman microscope as indicated in the description - a proportion of 100% or a proportion of ≥ 60% by volume and the target material consists furthermore of

   a. MoO₂ and Mo₄O₁₁ and also

   b. optionally one or more compound(s) from the group consisting of:

   - MoO₃,

   - further substoichiometric molybdenum oxides from the group consisting of
   Mo₁₇O₄₇, Mo₅O₁₄, Mo₈O₂₃, Mo₉O₂₆ and Mo₁₈O₅₂,

   - stoichiometric and/or substoichiometric Ta oxide(s),

   - stoichiometric and/or substoichiometric Nb oxide(s),

   - stoichiometric and/or substoichiometric V oxide(s), and

   - stoichiometric and/or substoichiometric Ti oxide(s),

   as one or more second phases which are distributed as islands in the matrix phase, with these second phases forming, based on a polished section of the target material and measured using a Raman microscope as indicated in the description, a total proportion of ≤ 40% by volume, wherein the proportion of the second phase MoOs in the target material is, based on a polished section of the target material - measured using a Raman microscope as indicated in the description - in the range 0 - 2% by volume, wherein the oxygen content in % by weight and determined by carrier gas hot extraction the target material is in the range from 26 to 32% by weight and wherein the total proportion of the doping element M in at.% in the target material relative to the total content of Mo and the doping element M in at.% is in the range 0.02 - 0.15.
2. Sputtering target according to claim 1, characterized in that, in the target material, the doping element M is formed by Ta and the matrix phase is formed by the mixed oxide (Mo_1-xTa_x)₆O₁₄ where 0.06 ≤ x ≤ 0.08.
3. Sputtering target according to claim 2, characterized in that the doping element Ta is entirely present as metal oxide in the target material, where the proportion of the phase Ta₂O₅ in the target material is, based on a polished section thereof and measured using a Raman microscope, ≤ 3% by volume.
4. Sputtering target according to any of claims 1 to 3, characterized in that, in the target material, the doping element M is formed by Nb and the matrix phase is formed by the mixed oxide (Mo_1-xNb_x)₅O₁₄ where 0.07 ≤ x ≤ 0.12.
5. Sputtering target according to claim 4, characterized in that the doping element Nb is entirely present as metal oxide in the target material, where the proportion of the phase Nb₂O₅ in the target material is, based on a polished section thereof and measured using a Raman microscope, ≤ 3% by volume.
6. Sputtering target according to any of the preceding claims, characterized in that the target material has a relative density of ≥ 95% and an electrical conductivity of ≥ 100 S/m at 25°C.
7. Sputtering target according to any of the preceding claims, characterized in that the target material is present as a powder-metallurgically produced solid body.
8. Sputtering target according to any of the preceding claims, characterized in that one or more second phases are present as finely and uniformly dispersed, small islands in the matrix phase in the target material, based on a polished section thereof and measured using a Raman microscope.
9. Use of a sputtering target according to any of claims 1 to 8 for vapour deposition of a molybdenum oxide-containing layer, characterized in that the sputtering process is carried out as DC sputtering process or pulsed DC sputtering process in a noble gas atmosphere without oxygen or alternatively with introduction of not more than 20% by volume of oxygen as reactive gas.

Images